Stroke recurrence after craniocervical artery dissection is strongly clustered in the hyperacute phase, with vascular morphology (specifically stenosis without dilatation and double lumen) rather than anatomical location independently associated with higher recurrence risk.
Key Findings
Results
CCAD was diagnosed in 0.94% of hospitalized patients with acute ischemic stroke or TIA, with intracranial dissection accounting for the majority of cases.
Among 75,903 hospitalized patients with acute ischemic stroke or TIA, 711 (0.94%) were diagnosed with CCAD.
Intracranial CCAD (I-CCAD) accounted for 76.1% of cases.
The mean age was 49.0 years, and 26.9% were women.
Median onset-to-arrival time was 7.7 hours, enabling early enrollment within 7 days of symptom onset.
Data were drawn from the nationwide CRCS-K-NIH registry in South Korea (2011–2021).
Results
Recurrent ischemic stroke after CCAD was heavily concentrated in the hyperacute phase, with the majority of 1-year recurrences occurring within the first week.
Recurrence was 8.7% on day 1, 17.4% within 1 week, and 19.9% at 1 year.
Incidence rates declined steeply from 271.2 events per 1,000 person-days on day 1 to 0.1 beyond 7 days.
The study analyzed recurrence across 4 discrete intervals: ≤1 day, 2–3 days, 4–7 days, and >7 days.
Recurrent events required MRI confirmation of new or extending infarction.
Results
Dissection anatomical location (intracranial vs. extracranial) was not associated with recurrence risk.
I-CCAD vs. E-CCAD location was evaluated as a predictor across all four time intervals.
No statistically significant association between dissection location and recurrence risk was identified.
This finding contrasts with vascular morphology features, which did show significant associations.
Results
Stenosis without dilatation was independently associated with higher 1-year recurrence risk after CCAD.
Adjusted hazard ratio (aHR) for stenosis without dilatation: 1.69 (95% CI 1.04–2.75).
This was identified as an angiographic morphological subtype evaluated alongside other vascular features.
The association was independent of other covariates in the multivariable model.
Results
The presence of a double lumen sign was independently associated with higher 1-year recurrence risk and demonstrated a distinct early temporal recurrence pattern specifically between days 2 and 3.
Adjusted hazard ratio (aHR) for double lumen: 1.87 (95% CI 1.15–3.04) for 1-year recurrence.
Double lumen sign was associated with recurrence between days 2 and 3 with an adjusted incidence rate ratio of 3.84 (95% CI 1.67–8.81).
This represents the strongest morphological predictor of recurrence identified in the study.
The double lumen sign was evaluated as one of the specific vascular imaging features alongside other angiographic subtypes.
Discussion
The study identified morphology-based, time-specific risk characterization as a framework to guide monitoring and secondary prevention after CCAD.
Early recurrence risk appeared to vary according to vascular morphology rather than anatomical location.
The findings support distinct risk windows that align with specific morphological features.
Limitations include the retrospective design and restriction to an East Asian population, which may limit generalizability.
The study enrolled patients within 7 days of symptom onset, addressing delayed enrollment limitations of prior research.
What This Means
This research suggests that when people have a stroke caused by a tear in one of the arteries supplying the brain (called craniocervical artery dissection, or CCAD), the risk of having another stroke is extremely high in the very first days after the initial event. In a large nationwide study from South Korea involving over 75,000 stroke patients, nearly 1 in 12 CCAD patients had a recurrent stroke on the very first day, and by the end of the first week, about 1 in 6 had experienced recurrence. After the first week, the risk dropped dramatically to very low levels. These findings highlight that the days immediately following a CCAD diagnosis are a critical window requiring close monitoring.
Importantly, this research suggests that where the artery tear is located in the body (inside vs. outside the skull) does not predict who will have a recurrent stroke. Instead, what the dissection looks like on imaging matters more. Specifically, two imaging features — a narrowed artery without ballooning (stenosis without dilatation) and the presence of a 'double lumen' (a visible false channel created by the tear) — were associated with significantly higher recurrence risk. The double lumen finding was particularly linked to recurrences occurring in the 2–3 day window after the initial event.
This research suggests that doctors may be able to use these specific imaging features to identify which patients are at highest risk for early recurrent stroke, potentially enabling more targeted monitoring and preventive treatments. The findings challenge the traditional approach of categorizing CCAD risk primarily by location and instead support a more nuanced, image-based approach tied to specific time windows. However, since this study was conducted in an East Asian population and used a retrospective design, further research in other populations would help confirm these findings.
Jang S, Hong J, Sohn S, Park H, Kang K, Youn M, et al.. (2026). Vascular Imaging Features and Time-Specific Recurrence Risk After Craniocervical Artery Dissection.. Neurology. https://doi.org/10.1212/WNL.0000000000218546